UC Santa Barbara

A real-time explosive gas sensing (DNT) in atmospheric pressure utilizing the noise

squeezing eect that occurs before a bifurcation event is investigated. A noise-squeezing

controller based on the statistics of phase noise is implemented using high-speed Lab-

VIEW eld programmable gated array. A high frequency TNT-molecularly imprinted

xed-xed microbeam sensor utilizes this nontraditional sensing strategy and performs

DNT sensing at various concentrations. Experiments are conducted using both noisebased

and sweep-based bifurcation tracking for a direct comparison. Results demonstrate

noise-based bifurcation tracking is not only capable of performing reliable frequency

tracking, but also show the method is superior to the bifurcation sweep-based

tracking. Over three orders of magnitude improvement in acquisition rate is achieved,

and as a result, condence and precision on bifurcation frequency estimation is signi-

cantly improved over the bifurcation sweep tracking method, enabling DNT sensing at

concentrations much below sub-ppb (parts-per-billion) level.